Tamm_BioBull

Reference: Biol. Bull. 229: 173–184. (October 2015)
© 2015 Marine Biological Laboratory
Functional Consequences of the Asymmetric
Architecture of the Ctenophore Statocyst
SIDNEY L. TAMM*
Bell Center, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
Abstract.
Ctenophores, or comb jellies, are geotactic
with a statocyst that controls the activity of the eight ciliary
comb rows. If a ctenophore is tilted or displaced from a
position of vertical balance, it rights itself by asymmetric
frequencies of beating on the uppermost and lowermost
comb rows, turning to swim up or down depending on its
mood. I recently discovered that the statocyst of ctenophores has an asymmetric architecture related to the sagittal
and tentacular planes along the oral-aboral axis. The four
groups of pacemaker balancer cilia are arranged in a rectangle along the tentacular plane, and support a superellipsoidal statolith elongated in the tentacular plane. By controlled tilting of immobilized ctenophores in either body
plane with video recording of activated comb rows, I found
that higher beat frequencies occurred in the sagittal than in
the tentacular plane at orthogonal orientations. Similar tilting experiments on isolated statocyst slices showed that
statolith displacement due to gravity and the resulting deflection of the mechanoresponsive balancers are greater in
the sagittal plane. Finally, tilting experiments on a mechanical model gave results similar to those of real statocysts,
indicating that the geometric asymmetries of statolith design
are sufficient to account for my findings. The asymmetric
architecture of the ctenophore statocyst thus has functional
consequences, but a possible adaptive value is not known.
vised structure affects its function. Specifically, does the
recently discovered asymmetric architecture of the ctenophore statocyst, with respect to the sagittal and tentacular
body planes (Tamm, 2014c), influence its functional responses during geotaxis?
Ctenophores, or comb jellies, are among the most beautiful of the marine zooplankton, with iridescent waves of
color coursing along their ciliary rows. They are the largest
animals to use cilia for locomotion, and the longest cilia in
nature are found in the eight rows of locomotory comb
plates of ctenophores. The large cilia in the comb plates and
in the macrocilia of ctenophores offer experimental advantages for studying motile and sensory functions of cilia
(Tamm, 1980, 1982, 2014a, b), and ciliary development and
regeneration (Tamm, 2012a, b).
Ctenophores have an equilibrium receptor system, or
statocyst, that uses the gravitational field as a reference
(Horridge, 1966, 1971, 1974; Budelmann, 1988). Ctenophores often assume a vertical position in the water with
their major longitudinal (oral-aboral) axis parallel to the
direction of gravity: either feeding mouth up at the surface, or swimming mouth downward, away from the
surface (Tamm, 1982, 2014a). If a ctenophore is tilted or
displaced from either position of vertical balance, it
rights itself by asymmetric frequencies of comb plate
beating on the uppermost and lowermost comb rows
(Horridge, 1974). The lower comb rows beat faster than
the upper rows, with ciliary power strokes directed aborally, when the animal turns to swim mouth upward
(negative geotaxis). The converse occurs during downward steering (positive geotaxis). After the animal regains a vertical orientation, all comb rows beat at a
similar frequency. The tendency to swim up or down (the
sign of geotaxis) is called a mood, and is influenced by
environmental and perhaps endogenous stimuli (Horridge, 1966, 1971, 1974; Tamm, 1982, 2014a).
Introduction
Francis Crick once said, “If you want to understand
function, study structure” (Crick, 1988). The statocyst of
ctenophores is a long-studied structure whose function is
generally understood. Recent findings, however, show that
the design of the structure is different from what was
supposed. This report investigates whether the newly reReceived 15 March 2015; accepted 28 May 2015.
* E-mail address: [email protected]
173
174
S. L. TAMM
The statocyst resides in the aboral sense organ, or apical
organ, at the aboral pole of the body. The ctenophore apical
organ is structurally and functionally unique in the animal
kingdom (Tamm, 2014a); and it contains other sensory
elements besides the statocyst, including putative photoreceptor cells that express opsins (Schnitzler et al., 2012) and
pressure receptors (Tamm, 1982), with possible influences
on geotactic mood. The single large statolith consists of
numerous aggregated living cells, or lithocytes. Each lithocyte is filled with a membrane-bound calcareous concretion
surrounded by a rim of cytoplasm and nucleus (Samassa,
1892; Krisch, 1973; Aronova, 1974; Tamm, 1982, 2014a, c;
Noda and Tamm, 2014). The statolith is supported upon the
ends of four sickle-shaped groups of cilia called balancers,
each located in a body quadrant and comprising 150 –200
mechanoresponsive cilia. The balancers are motile and act
as pacemakers for beating of the comb rows (Chun, 1880;
Horridge, 1966, 1971, 1974; Tamm, 1982, 2014a). Each
beat of a balancer is propagated mechanically as a wave of
beating along its two narrow tracts of short cilia (ciliated
grooves) to a pair of comb rows in that quadrant of the body.
The beat frequency of a balancer depends on the amount it
is bent by the gravitational load of the statolith and the
direction of bending (Tamm, 1982, 2014a). The same mechanical stimulus can either excite or inhibit beating of a
balancer and the two comb rows connected to it, depending
on the ctenophore’s sign of geotaxis, or mood (Tamm, 1982,
2014a; Lowe, 1997). The balancers thus serve as both
mechanoreceptors and direct effectors to control swimming
direction and orientation with respect to gravity (Horridge,
1966, 1971, 1974; Tamm, 1982, 2014a).
Ctenophores display features of bilateral symmetry
around the oral-aboral axis: the flattened stomodeum defines
the sagittal plane, and an orthogonal tentacular plane passes
through the two tentacle pouches. Lobate ctenophores such
as Mnemiopsis are compressed in the tentacular plane and
expanded in the sagittal plane, with two oral lobes on either
side of the mouth (Fig. 1). The four comb rows bordering
the sagittal plane (subsagittal rows) are longer and have
more comb plates than the four shorter comb rows bordering the tentacular plane (subtentacular rows).
I recently discovered that the statolith of Mnemiopsis,
Pleurobrachia, and Beroe is not a spherical mass of lithocytes, as previously supposed, but has a superellipsoidal
shape with its major axis parallel to the tentacular plane
(Fig. 1) (Tamm, 2014c). This non-radial symmetry of the
statolith is due to two factors: a rectangular arrangement of
the four balancers, which arise at the vertices of a rectangle
with a 3:1 length-width ratio oriented along the tentacular
plane (Fig. 1) (Tamm, 2014c); and the addition of new
lithocytes only at the two ends of the statolith by transport
along the ciliary surfaces of the balancers (Noda and Tamm,
2014).
Figure 1. Diagram of Mnemiopsis (top panels) and higher magnification
of its statocyst (lower panels), as viewed from the side in the sagittal plane
(S, left column) and in the tentacular plane (T, right column). The statocyst
is located within the small open circle near the aboral end of the animal
(top). Two rounded lobes (L) project from the oral end. The longer
subsagittal comb rows (ss) and the shorter subtentacular comb rows (st) are
shown by heavy lines running parallel to the aboral-oral axis (vertical). The
major (a), minor (b), and depth (c) axes of the superellipsoidal statolith
(black) and the rectangular arrangement of the four balancers (ba) on the
epithelial floor (stippled) are oriented orthogonally in the two body planes.
Note that the longer a-axis of the statolith and the greater spacing between
balancer pairs occurs along the tentacular plane. (Modified from Tamm,
2014c).
My aim was to discover whether these bilateral asymmetries in statocyst architecture affect the function of the
statocyst during geotactic behavior of ctenophores.
Materials and Methods
Organisms
Mnemiopsis leidyi A. Agassiz, 1865, were carefully
dipped from the surface of the sea at Woods Hole, Massachusetts. Freshly collected animals were used immediately
or kept briefly in flowing seawater at the Marine Biological
Laboratory (Woods Hole, MA).
STATOCYST FUNCTION IN CTENOPHORES
Whole-animal tilting
Adult Mnemiopsis (3.5 cm average body length along the
oral-aboral axis) were individually mounted with two insect
pins on a 9-cm diameter transparent Sylgard 184 (Dow
Corning) disk. The center of the disk was attached to a glass
rod axle inserted through a Lucite support tray. The disk
could be manually rotated in a vertical plane to different
orientations with respect to gravity in a Lucite viewing box
filled with seawater (cteno-tilter; Fig. 2) (Tamm, 1980,
1982). Each animal was pinned first in either the sagittal or
tentacular body plane. The beat frequencies of comb rows
on opposite sides of the body, viewed in profile from the
front of the cteno-tilter, were recorded by a video camera at
four orientations to gravity in the following sequence: the
oral-aboral axis vertical with the aboral pole facing up (0°),
the oral-aboral axis horizontal with the aboral pole facing to
the right (90°), the oral-aboral axis vertical with the aboral
pole facing down (180°), and the oral-aboral axis horizontal
with the aboral pole facing to the left (270°).
Two horizontal tilt positions, at 90° and 270°, were
chosen because the differences in beat frequencies between
the uppermost and lowermost comb rows are graded with
tilt angle and reach a maximum at 45° through 90° deviation
from vertical (Tamm, 1980, 1982). Right-angle orientations
of the animal also were more convenient to select.
After testing in a given body plane (sagittal or tentacular),
each animal was unpinned and remounted on the disk in the
other mutually perpendicular plane (tentacular or sagittal)
for a second round of testing at the same orientations (0°,
90°, 180°, and 270°). The body plane used first for testing
was alternated in successive animals to control for possible
effects on beat frequencies of time and repeated pinning in
the cteno-tilter.
The activity of comb rows on opposite sides of the body
was recorded for 1–1.5 min at each body orientation. This
cycle of tilting and recording was usually repeated, at least
for the two horizontal orientations at 90° and 270°. Activation or inhibition of beating at a given orientation was often
uniform during the recording period, but sometimes included temporary slowdowns or pauses. Measurements of
beat frequency (see next section) were made only during
continuous activity.
Beat frequency recording
The activity of comb rows of pinned animals was recorded by a Zeiss ZVS-3C75DE video camera and Nikon
Micro-NIKKOR 55 mm 1:2.8 lens aimed at the front wall of
the transparent, back-illuminated cteno-tilter (normal to the
Sylgard disk and plane of tilting) (Fig. 2). A camera exposure time of 0.5 ms or 1 ms gave sharp, still-field images of
successive stages in the comb plate beat cycle. Images and
audio tracks of relevant information were recorded on SVHS videocassettes with a Panasonic AG-7355 VCR that
175
allowed field-by-field playback (16.7 ms per field), coupled
to a QSI-VFF 6030 field/frame counter and video monitor.
Beat frequency analysis
Changes in geotactic sign are often accompanied by
variations in beat frequencies and transient irregular activities of the imaged comb rows. Therefore, a main criterion
for selecting animals to analyze was a consistent sign of
geotaxis (up or down mood, not switching) during testing at
all four orientations to gravity in both body planes of tilting
(sagittal and tentacular).
Beat frequency of a comb row was determined by following the upright stage of the beat cycle (midway through
the effective stroke) of a single comb plate (Fig. 2). The
number of video fields required for 10 successive beats of
the selected plate was counted. Slight variations between the
actual beat rate and the fixed video field rate were usually
compensated for over 10 successive beats, with a maximum
possible counting error of one video field.
Independent counts (usually 7–10) of the number of fields
for 10 beats were made at approximately regular intervals
during the 1–1.5-min observation period at each tilt orientation. In order to maximize the response for all animals, the
five “fastest” counts (smallest number of fields per 10
beats ⫽ highest frequencies) were selected for conversion to
Hz and calculation of the mean beat frequency and standard
deviation at each tilt orientation for sagittal versus tentacular planes of tilting. In all cases, a possible one-field error in
field counts was less than the standard deviations of mean
beat frequencies.
Isolated statocyst tilting
Adult animals (4 – 4.5 cm along the aboral-oral axis) were
placed individually in a Sylgard-floored glass fingerbowl of
seawater on a transmitted light stage and pinned in the
sagittal plane. Two parallel transverse sections across the
body were cut with a dissecting scissors aboral and oral to
the statocyst region. The excised section was transferred
into a small Sylgard dish and oriented with the aboral
surface up for further dissection under a stereo microscope.
A 2–3 mm-thick flat slice containing the statocyst was cut
and trimmed with dissecting and iridectomy scissors along
the sagittal or the tentacular plane, depending on the desired
plane of tilting of the statocyst. Sharp differential interference contrast (DIC) imaging of the statolith and balancers
was difficult to achieve, even when one side of the slice (the
future upper surface on the slide) was cut very close to the
statocyst to remove obscuring tissue.
The tissue slice was transferred into one side of a doublechamber microscope slide ridged with Dow Corning vacuum grease. This chamber was connected to the other (overflow) chamber by a short segment of glass capillary. A 22 ⫻
176
S. L. TAMM
Figure 2. Motor responses of comb rows in a negatively geotactic Mnemiopsis in the cteno-tilter. The animal
is pinned to a Sylgard disk and rotated in a vertical plane under seawater (see Materials and Methods). The end
of the glass rod axle in the center of the disk is visible behind the animal. The location of the statocyst near the
aboral end is indicated by an open circle. The mouth and oral lobes are outside the field. The animal is pinned
in the sagittal plane in the left half (column) at 90° (A) and at 270° (C). The longer subsagittal comb rows (ss)
are in profile view, and the lowermost rows beat actively in metachronal waves, which travel orally. Arrowheads
mark single comb plates performing an effective stroke toward the aboral end. Comb plates in the uppermost
subsagittal rows do not beat, but lie at rest pointing orally. In the right half (column), the same animal is pinned
in the tentacular plane at 90° (B) and 270° (D). The shorter subtentacular comb rows (st) are in profile view, and
the lowermost rows beat in metachronal waves traveling orally. Arrowheads indicate comb plates in an effective
stroke. The uppermost subtentacular rows are inactive. This method was used to compare differences between
beat frequencies of the activated lowermost comb rows in the same animal when tilted in the sagittal or tentacular
plane at 90° (A vs. B) and at 270° (C vs. D). Scale bar, 0.25 cm.
22-mm coverslip was lowered over the flat upper side of the
tissue slice in the first well and gently compressed to seal
the preparation and expel excess seawater into the open
overflow chamber from which it was removed. The leakproof slide was then mounted on a vertical rotating stage of
a horizontal Zeiss Universal microscope (micro-tilter;
Tamm, 1980, 1982). An isolated living statocyst could be
tilted in any orientation with respect to the direction of
gravity in the sagittal or the tentacular plane. However,
unlike the experiments with whole animals in a cteno-tilter,
the body plane in which an isolated statocyst was tilted
could not be switched, but was determined by the orientation of the dissection.
A 16⫻/0.35 DIC objective with the same video equipment used for whole-animal tilting was used to image
micro-tilting of isolated statocysts.
Analysis
Changes in gravity-dependent displacements of the statolith and resulting deflections of the balancers at different
orientations of the statocyst were determined by tracings
177
STATOCYST FUNCTION IN CTENOPHORES
from a monitor and video prints of selected still-field images. At vertical orientations (0°, 180°), the center of the
statolith lies on the oral-aboral axis on the midline of the
statocyst cavity, halfway between opposing balancers. At
the two horizontal orientations (90°, 270°), the downward
displacement of the statolith was measured relative to this
midline reference, which was drawn halfway between the
bases of the upper and lower balancers and normal to
the epithelial floor for either body plane. The length of the
statolith’s major a-axis (if tilted in the tentacular plane) or
minor b-axis (if tilted in the sagittal plane) was measured,
and that part of the axis below the midline—which never
exceeded either axis—was divided by the full length of the
major or minor axis to give the relative (percent) downward
displacement of the statolith due to gravity. Since at a
vertical position without lateral displacement (0° or 180°),
50% of the length of the major or minor axis of the statolith
lies on either side of the midline, a null value of 50% was
subtracted from the observed percent downward displacement of the statolith. The percent downward displacement
of the statolith at 90° and 270° for sagittally and tentacularly
tilted statocysts was averaged and plotted as a function of
the plane of tilting.
The deflections of lower balancers due to downward
statolith displacement at 90° and 270° orientations were
quantified and compared in the sagittal versus tentacular
planes of tilt by measuring the downward distance from the
base of a balancer to its distal insertion into the statolith.
This method is independent of the approximately three
times greater spacing between upper and lower balancers in
the tentacular versus sagittal plane.
Statistical tests
The statistical significance of differences between sample
means was evaluated in two ways. Differing degrees or
classes of significance were estimated in populations of
animals whose mean beat frequencies of the lowermost
rows were higher, but by considerably different amounts, in
either the sagittal or the tentacular plane. The standard
deviation for the distribution of sample means, or the standard error of the difference between the means, was used to
compare relative significances. A difference of three or
more standard errors between the mean beat frequencies for
each plane was classed as definitely significant; a difference
of more than two standard errors was regarded as probably
significant; and a lesser difference, as not significant (Moroney, 1956). This treatment provided a more informative
and complete account of the relative significance of the data
than a “significant or not significant” test using a P-value.
Conventional two-tailed t-tests were used for other statistical evaluations.
Mechanical model
A weight-and-springs mechanical simulation of the statocyst was built to help interpret the data on living statocysts.
The statolith is represented by a 6.3-cm-long ovoidal piece
of Styrofoam sprayed with small, 2–3 mm Styrofoam beads.
The beads somewhat resembled the aggregated lithocytes of
living statoliths, but were much smaller relative to statolith
size. The mass (load) of the lightweight statolith was set by
nails placed in the center along the major axis.
The four balancers were represented by wire coil springs.
The ends of the springs were attached to the statolith and to
a flat base representing the epithelial floor of the statocyst.
The geometric positions of spring attachment to both the
statolith and the base corresponded to the structural asymmetry of the living statocyst with respect to sagittal and
tentacular body planes. The model was tilted at different
orientations to gravity in the sagittal and tentacular planes
and photographed with an Olympus FE-170 digital camera.
Results
Whole-animal tilting
All of the animals that displayed a consistent sign of
geotaxis during the tilting experiments and were used for
analysis of beat frequencies were negatively geotactic. This
finding is not surprising, since ctenophores were collected at
the water surface during fairly calm weather, and were
predominantly in an up mood when selected for this study.
At the initial orientation (0°) of a negatively geotactic
Mnemiopsis in the cteno-tilter with the oral-aboral axis
vertical and the aboral pole (statocyst) facing up, only
occasional waves of beating were seen on all eight comb
rows, as described previously for Pleurobrachia and Beroe
in an up mood (Tamm, 1980, 1982). Rotating the animal
clockwise to 90° with the oral-aboral axis horizontal and the
aboral pole facing to the right activates fast beating of only
the lowermost four comb rows (Fig. 2, top row). Since the
power stroke of comb plates is directed aborally (in the
opposite direction to the antiplectic metachronal waves),
this asymmetric pattern of activity of the lower versus upper
rows would cause a free animal to turn with its mouth facing
upward (Tamm, 1980, 1982, 2014a). Turning the animal to
180° with the oral-aboral axis vertical and the aboral pole
facing down results in fast beating of all rows (not shown),
which would cause a free animal to swim upward. Finally,
at the 270° orientation with the oral-aboral axis horizontal
and the aboral pole facing to the left, only the lowermost
four rows (formerly the uppermost rows at 90°) beat fast
and the uppermost rows are inhibited (Fig. 2, bottom row),
again causing a free animal to turn mouth upward.
This pattern of selective, gravity-dependent excitation
and inhibition of comb row beating is characteristic of
negative geotaxis (Tamm, 1980, 1982, 2014a), and occurs
178
S. L. TAMM
regardless of the body plane (sagittal or tentacular) in which
an animal is tilted (Fig. 2). For animals pinned in the sagittal
plane, comb plates in the longer subsagittal rows were in
profile view and their beat frequencies were recorded (Figs.
1, 2). For animals pinned in the tentacular plane, comb
plates in the shorter subtentacular rows were in profile view
so their beat frequencies were recorded (Figs. 1, 2). Since
each balancer controls the beating of one subsagittal row
and one subtentacular row, the motile responses of all four
balancers to the direction of gravity, as expressed by comb
row activity, were evident regardless of the plane of tilting.
The beat frequency activation response of the lowermost
comb rows at 90° and 270° body orientations was therefore
compared as a function of the plane, sagittal or tentacular, in
which the animal was tilted. Theoretically, the beat frequencies of the lowermost rows at 90° and 270° should be
similar (Tamm, 1982). Both orthogonal orientations to gravity were used as controls for possible internal variations or
abnormalities in individual animals, and were then discarded.
There was no significant effect on recorded beat frequencies related to which body plane was tested first in the
cteno-tilter. Therefore, the data for both orders of testing
were combined into one set for tilting in the sagittal plane or
the tentacular plane.
At a 90° body orientation, the mean beat frequency of the
lowermost comb rows was higher when an animal was tilted
in the sagittal plane in 17 of the 21 animals analyzed (81%)
(Fig. 3). However, there were considerable variations from
animal to animal between the sample means in the two
planes. The number of standard errors of the difference
between means was used to judge their relative significances (see Materials and Methods). In the 17 animals noted
above, the difference between the mean beat frequency for
each plane was definitely significant in 9 animals, probably
significant in 2 animals, and not significant in 6 animals
(Fig. 3). At the same 90° orientation, the mean beat frequency of the lowermost rows was higher in the tentacular
plane in 4 of the 21 animals (19%). In these animals, the
difference between the mean beat frequency for each plane
was probably significant in 2 animals, but not significant in
the other 2 animals (Fig. 3).
At the other orthogonal orientation (270°), a similar but
less pronounced pattern was found. The mean beat frequency of the lowermost rows was higher when an animal
was tilted in the sagittal plane for 15 of the 21 animals
(71%). In these animals, the difference between the mean
beat frequency for each plane was definitely significant in 4
animals, probably significant in 1 animal, and not significant
in 10 animals (Fig. 3). At the same 270° orientation, the
mean beat frequency of the lowermost rows was higher in
the tentacular plane in 6 of the 21 animals (29%). In these
Figure 3. Comparative distribution of numbers of animals whose mean
beat frequencies of the lowermost comb rows were higher when an animal
was tilted in the sagittal plane than in the tentacular plane (S ⬎ T) or vice
versa (T ⬎ S) at 90° (left column pair) and 270° (right column pair).
Twenty-one animals were tested at the two orthogonal orientations in both
body planes. The total number of S ⬎ T animals and T ⬎ S animals in each
column was subdivided into the number of animals in three different
significance classes: definitely significant (black), probably significant
(gray), and not significant (white) (based on standard errors of the difference; see Materials and Methods).
animals, the difference between the means was definitely
significant in 1 animal, probably significant in 2 animals,
but not significant in the remaining 3 animals (Fig. 3).
In the smaller number of cases where the mean beat
frequency of the lowermost rows was higher in the tentacular plane at 90° or 270°, the mean beat frequency of the
lowermost rows at the other orthogonal orientation of the
same animal (270° or 90°) was higher in the sagittal plane
in all but one animal, in which the mean beat frequency of
the lowermost comb rows of that animal was higher in the
tentacular plane at both 90° and 270° orientations.
Figure 4A–C shows three examples at 90° orientation
of significantly higher mean beat frequencies of the lowermost rows when animals were tilted in the sagittal
plane (taken from the 9 cases in Fig. 3). In Figure 4A, the
mean beat frequency in the sagittal plane was 24 Hz
compared with 18 Hz in the tentacular plane, a 33%
difference. Figure 4D shows significantly higher mean
beat frequencies of the lowermost rows in the sagittal
plane at both 90° and 270° orientations of the same
animal, with similar beat frequencies noted in either body
STATOCYST FUNCTION IN CTENOPHORES
Figure 4. Individual examples of significantly higher mean beat frequencies of the lowermost rows in the sagittal plane (gray) and the
tentacular plane (white) at 90°, for four different animals (A–D), including
one animal at both 90° and 270° orientations (D). Note that in this animal,
the higher beat frequencies in the sagittal plane are similar at both 90° and
270°, as are the lower beat frequencies in the tentacular plane at these
orientations (D). Error bars are standard deviations.
plane at both orthogonal orientations. For all sagittal
versus tentacular plane pairs shown, P ⬍ 0.01 except at
270° in Figure 4D, where P ⬍ 0.05.
Isolated statocyst tilting
To interpret the differences in comb row activity at 90°
and 270° for animals tilted in the sagittal or the tentacular
plane, I examined the relative downward displacement of
the statolith and resulting deflections of the mechanoresponsive balancers with respect to the orthogonal body planes.
Tissue slices containing the statocyst were dissected in the
sagittal or the tentacular plane, mounted on leakproof slides,
and rotated to different orientations with respect to gravity
on a vertical rotating stage of a horizontal microscope
(micro-tilter). Due to variations in dissection and tissue
orientation on the slide, many preparations did not allow
satisfactory imaging of the statolith and balancers, and were
discarded. Of the slides used, the balancers nearest the
sagittal or tentacular surface of the slice were the clearest.
Because the statolith and balancers reside in slightly different focal planes when viewed laterally, both structures usually were not completely in focus, and required throughfocus imaging for analysis.
At the two vertical orientations to gravity of isolated
statocyst preparations, corresponding to 0° and 180° orientations of whole animals in a cteno-tilter, statolith position
and balancer configurations were oppositely and symmetrically affected by gravity, and in a similar manner in both the
sagittal and tentacular planes (Fig. 5).
The superellipsoidal statolith was located with its center on the oral-aboral axis at the intersection of the
sagittal and tentacular planes (i.e., on the midline of the
epithelial floor in either plane; Fig. 5). At these orientations, all four balancers experience equal loading or
179
unloading by the statolith, and usually show the same
beat frequency, which depends on the animal’s mood (see
Tamm, 1980, 1982).
At 0° (mouth downward, statocyst upward in intact animals), the statolith rested close to the epithelial floor,
against the bent bases of the balancers (Fig. 5A, B). Since
only the basal part of a balancer actively moves during a
beat (Tamm, 1980, 1982), the beating of balancers at 0°
transiently lifts the statolith slightly away from the epithelial
floor. The distal regions of the balancers were closely
curved around the sides of the statolith near their entrance
into it. At 180° (mouth upward, statocyst downward in
intact animals), the statolith hung some distance down from
the epithelial floor at the straightened distal ends of the
balancers (Fig. 5C, D).
At the two horizontal orientations to gravity of an isolated
statocyst, corresponding to 90° and 270° positions of whole
animals in a cteno-tilter, the statolith was displaced downward from its equilibrium position at the center of the
statocyst cavity (Fig. 6). The lower (activated) balancer was
pushed away from the midline by the load of the statolith,
and the upper (inhibited) balancer was pulled toward the
midline (Fig. 6). In both cases, the weight of the statolith
bent only the curved distal region of the balancers, but their
motile responses occurred at the base (Tamm, 1980, 1982).
Figure 7 illustrates the geometric methods used for quantitating and comparing statolith displacements and the resulting balancer deflections at orthogonal orientations (270°
shown) in both sagittal and tentacular planes (see Materials
and Methods).
Measurements of isolated statocysts in slices prepared in
the sagittal plane (n ⫽ 11) or in the tentacular plane (n ⫽ 9)
show that the mean statolith displacement from the midline
at 90° is similar to that at 270° in either body plane.
However, the mean statolith displacements at 90° and 270°
are about 5% greater in statocysts tilted in the sagittal plane
(⬃30%) than in the tentacular plane (⬃25%) (Fig. 8). This
difference between statolith displacements in the two body
planes is significant by three standard errors with a
P-value ⬍ 0.01.
The downward deflections of the lowermost balancers at
90° and 270° were analyzed because intact animals in the
cteno-tilter were negatively geotactic with fast beating of
the lowermost comb rows at these orientations. Since these
balancers were more clearly imaged at 270° than at 90° tilt,
the deflections at 270° orientation were analyzed in the two
body planes (Fig. 7). The mean downward deflection of the
distal region of a lower balancer at 270° was 42 ⫾ 3.8 ␮m
(n ⫽ 5) in the sagittal plane and 28 ⫾ 3.0 ␮m (n ⫽ 8) in the
tentacular plane (P ⬍ 0.01).
Thus, at a right-angle orientation to gravity, an approximately 1.2 times greater downward displacement of the
statolith in the sagittal than in the tentacular plane is
180
S. L. TAMM
Figure 5. Micro-tilting of isolated statocysts at vertical orientations in the sagittal versus tentacular planes at
0° (A vs. B) and 180° (C vs. D). The four images are arranged with sagittal views in the left column and
tentacular views in the right column, with the same orientation to gravity on the left and right for comparisons
between the two planes. Note that the bases of the balancers (b) are set closer together in the sagittal plane
(A, C) than in the tentacular plane (B, D) at both orientations. In B, the compressed balancers are partially
obscured. The two orientations in the sagittal plane are from the same animal, and those in the tentacular plane
are from a different animal. Scale bar, 10 ␮m.
correlated with about a 1.5 times greater downward deflection of the lowermost balancers between these two
planes.
The sign of geotaxis of isolated statocyst slices in the
micro-tilter was more variable within a preparation and
between different slides compared with intact animals
used in the cteno-tilter, perhaps due to dissection, absence of the rest of the animal’s body, microscopy conditions, or other factors. Because of these variations, the
beat frequency responses of balancers in isolated statocysts could not reliably be compared between the sagittal
and tentacular planes.
Mechanical model
The model statocyst in a vertical orientation to gravity (0°
in living statocysts) presented its Styrofoam statolith with
its center on the midlines of the sagittal and tentacular
planes, and with similar configurations of all four supporting balancer springs (Fig. 9A). At a horizontal orientation of
the model (270° in living statocysts), the statolith was
displaced downward by gravity a greater distance from the
midline when tilted in the sagittal plane than in the tentacular plane (Fig. 9B vs. C). This result was due entirely to
geometry, that is, the elongated distribution of load supported on a rectangular arrangement of flexible cantilevers.
The degree of statolith displacement in each plane was
remarkably similar to that found in living, isolated statocysts (Figs. 6, 8), due to trial-and-error selection of an
appropriate weight for a statolith resting on the wire coil
springs used in the model. The deflection of the lowermost
balancer springs was therefore greater when the model was
tilted in the sagittal plane (Fig. 9B, C), similar to the
findings in living statocysts (Fig. 6). The significance of the
model for understanding the behavior of the living statocyst
is discussed below.
STATOCYST FUNCTION IN CTENOPHORES
181
Figure 6. Micro-tilting of isolated statocysts at horizontal orientations in the sagittal and tentacular planes at
90° (A vs. B) and 270° (C vs. D). The four images are arranged with sagittal views in the left column and
tentacular views in the right column, with the same orientation to gravity on the left and right for comparisons
between the two planes. Note that the bases of the balancers (b) are closer together in the sagittal plane (A, C)
than in the tentacular plane (B, D) at both orientations. The two orientations in the sagittal plane are from the
same animal, as in Fig. 5A, C, and those in the tentacular plane are from the same animal, as in Fig. 5B, D.
Magnification is the same as in Figure 5.
Discussion
This study reports the functional consequences of a recently discovered design asymmetry in the geotactic organ
of ctenophores (Tamm, 2014c). The architecture of the
ctenophore statocyst previously had been considered to be
radially symmetric about the longitudinal body axis (oralaboral axis) and in the two orthogonal planes of symmetry
(sagittal and tentacular) along this axis. For this reason,
previous studies on the mechanism of geotaxis, particularly
the beat frequency responses of specific comb rows to
angular changes in body orientation to gravity, as seen in a
cteno-tilter (Tamm, 1980, 1982, 2014), did not take into
account (or alter) the body plane in which an animal was
tilted. For example, cydippids like Pleurobrachia, the most
radially symmetric ctenophores with nearly equal sagittal
and tentacular axes, were routinely tested in the tentacular
plane. In comparison, beroids (Beroe) and lobates (Bolinopsis) were commonly tilted in their flattened sagittal or
stomodeal plane; the results in both planes were treated
equivalently (Tamm, 1980, 1982, 2014a).
The discovery of a bilateral morphological asymmetry of
the statocyst with respect to the sagittal and tentacular
planes (Tamm 2014c) prompted the present investigation
into whether this newly revised structural plan affects the
geotactic responses of ctenophores. I report here that it does:
function follows structure, as Crick (1988) suggested, also
in the statocyst of ctenophores. The cteno-tilting experiments on whole animals in an up mood convincingly show
that the activated lowermost comb rows at orthogonal orientations to gravity beat faster when an animal is tilted in
the sagittal plane than in the tentacular plane (Fig. 3).
Although the differences in mean beat frequencies between
the two body planes of an animal varied widely and ranged
from insignificant to highly significant, a trend for an increased motile response of the lowermost comb rows in the
sagittal plane was clear (Figs. 3, 4). Although animals in a
down mood were not available for testing, the same results
182
S. L. TAMM
Figure 7. Protocol for determining statolith displacements and balancer
deflections due to gravity in isolated statocysts at orthogonal orientations
(270° shown). In the sagittal plane (S), the minor b-axis of the superellipsoidal statolith is parallel to the gravity vector (down) and the balancers are
closely spaced. In the tentacular plane (T), the major a-axis of the statolith
is parallel to gravity and the balancers are farther apart (see Figs. 1, 5, 6).
The relative downward displacement of the statolith from the midline (m)
of the epithelial floor (f) was calculated as the length of the axis below the
midline (x) divided by the full length of the axis (b or a) as a percent, and
subtracting 50% (the equilibrium value at vertical orientations). The downward deflection of a lowermost balancer (black) at 90° and 270° was
measured as the vertical distance along the floor from its base to its
entrance into the statolith (dashed lines and arrow). (See Materials and
Methods).
are assumed for positively geotactic animals, except that the
uppermost, not lowermost, comb rows would be activated at
90° and 270° orientations (see Tamm, 1980, 1982, 2014a).
Micro-tilting of isolated statocysts in tissue slices confirmed expectations from whole-animal tilting: that both the
amount of downward displacement of the statolith and the
degree of deflection of the lowermost balancers by the load
of the statolith are greater in the sagittal plane. Thus, the
differences in comb row beat frequencies between the two
body planes is correlated with a greater gravitational stimulus of the sensory-effector organ (statocyst) controlling
geotaxis.
Figure 8. Relative statolith displacements at 90° and 270° in isolated
statocysts tilted in sagittal (dark) or tentacular (white) planes. Y-bars are
standard deviations.
The approximately 1.2 times greater downward displacement of the statolith in the sagittal plane results in about a 1.5
times greater downward deflection of the lowermost balancers
in the sagittal plane, implying a non-linear stress-strain relationship between applied load and ciliary deformation. Previous micromanipulation experiments showed that the beat frequency of isolated, statolith-free balancers of Mnemiopsis
increased in a linear fashion with increasing amounts of deflection by a suction-pipette attached to the ciliary tips (Lowe,
1997). Since I was not able to quantify the effects of differing
degrees of statolith displacement and balancer deflection on the
beat frequency of a balancer, the stimulus-response relationship of balancers in situ remains a topic for future work.
However, it seems clear from the whole-animal experiments
that the greater downward deflection of the lowermost balancers in isolated statocysts tilted in the sagittal plane causes a
higher beat frequency of these balancers with correspondingly
higher rates of beating of the lowermost comb rows, as observed in the sagittal plane using the cteno-tilter.
Explanation of results
The similarities between the mechanical model (Fig. 9) and
living statocysts in the different amounts of statolith displacement and balancer deflection when tilted in sagittal versus
tentacular planes offers important clues for interpreting the
results.
The wire helical springs (balancers) have a similar resistance to lateral flexion regardless of the direction of applied
force (i.e., statolith load in the sagittal or tentacular plane).
The observed differences in statolith displacement and balancer deflections between the two body planes in the model
thus reflect the physical asymmetries of its design: four
cantilever balancers are located at the vertices of a rectangle
lying in the tentacular plane, which are end-loaded by an
ovoidal gravitational mass with its long axis also oriented in
the tentacular plane. This matching geometric asymmetry of
load and supporting elements is sufficient by itself to account for the difference in mechanical and motile outputs
observed between tilting living statocysts and ctenophores
in sagittal versus tentacular planes.
Of course, a model (or theory) that is consistent with the
data does not necessarily mean it is correct, or even a
complete explanation. The most important simplification of
the model is the nature of the balancers. Living balancers,
unlike coil springs, are not cylindrical in shape with a
uniform diameter (even though balancers are commonly
depicted as such in diagrams like Fig. 1). Instead, the base
of a balancer is flared into a wide V, with the cilia arranged
in rows forming a rhomboidal lattice (Tamm, 1982; Tamm
and Tamm, 2002). About one third of the 150 –200 cilia in
a balancer diverge from the main group near the base and
overlap the bent distal ends of cilia at the beginning of a
subsagittal ciliated groove (Tamm, 1982). The majority of
STATOCYST FUNCTION IN CTENOPHORES
183
Figure 9. Mechanical model of the statocyst viewed at 0° in the tentacular plane (A), tilted to 270° in the
sagittal plane (B), and tilted to 270° in the tentacular plane (C). At 0°, the ovoidal, beaded Styrofoam statolith
and the four balancer springs are symmetrically disposed about the midlines (black lines) of the sagittal (S) and
tentacular (T) planes (see Fig. 5). Note that the four balancers are arranged in a rectangle parallel to the tentacular
plane and the longer axis of the statolith (see Fig. 1). At 270°, the statolith is displaced farther downward from
the midline in the sagittal plane (B) than in the tentacular plane (C). As a result, the lowermost balancers are
deflected a greater distance from the midline in the sagittal plane (see Fig. 6).
the balancer cilia that curve upward toward the statolith
appear differently in the two body planes: as a broad ribbon
when viewed in the sagittal plane, and as a narrow, tapering
stalk when observed in the tentacular plane (Figs. 5, 6, 7).
Therefore, the balancer shaft, which is deflected under load,
is not cylindrical like a coil spring, but is shaped like a strip
flattened in the sagittal plane and has a different crosssectional profile in each body plane. However, the midlines
of the V-shaped bases of the balancers are not oriented
preferentially with respect to either the sagittal or tentacular
plane, but lie diagonally to each plane; that is, their orientation is geometrically equivalent in both body planes.
Any or all of these known features of balancer structure
and orientation might affect their deformation under a load
directed in a given plane. Future work should elucidate the
importance of the geometric asymmetries of load and balancer arrangement, as exemplified by the model, in relation
to the structural properties of real balancers for understanding the observations reported here.
Possible adaptive value
Since the observed differences in ciliary beat frequencies
are likely to arise largely from the asymmetric design of the
statocyst, one may ask first whether this architectural constraint is itself an adaptation for an essential function, or
184
S. L. TAMM
exists only as a byproduct of other processes or factors
(Gould and Lewontin, 1979). If the latter is the case, then
these functional findings may not have any adaptive value in
the life of ctenophores. A more relevant question is whether,
in free, unrestrained animals turning to swim up or down,
the effectiveness of the righting response depends on the
plane in which an animal is tilted. If the righting response of
animals in nature is indeed stronger in the sagittal than the
tentacular plane, then one may ask if this behavior is advantageous for ctenophores, and, if so, how.
One approach to this question is to try to rule out possible
candidates, such as a benefit related to asymmetric body
shape. In lobate ctenophores like Mnemiopsis, the body is
flattened in the tentacular plane and expanded along the
sagittal plane (Fig. 1). For Mnemiopsis in a vertical position
in the water (at 0° or 180°), an approaching directional
disturbance (wave or current) is more likely to have a
greater impact encountering the broader sagittal surface
than the narrower tentacular side, thereby causing an animal
to be tilted more often and to a greater extent in the tentacular plane. A stronger righting response in the tentacular
plane would therefore be expected to be more advantageous
for regaining vertical balance. But the opposite response
was found, implying that these results would be disadvantageous by this hypothesis.
A further argument against a body shape-related, adaptive value is the well-known diversity of body plan symmetry in different orders of ctenophores. Cydippids, like Pleurobrachia, are almost radially symmetric, with nearly equal
sagittal and tentacular axes. Yet their statocyst shares the
same structural asymmetries with respect to the two body
planes as the statocyst of Mnemiopsis (Tamm, 2014c). If
function follows structure in the statocyst of different ctenophores, I would expect to find a higher beat frequency in the
sagittal plane of tilted Pleurobrachia, as reported here for
Mnemiopsis. This theory has not been tested.
In conclusion, testable predictions from other proposed
adaptive “reasons” for my findings clearly await better
hypotheses based on a more extensive knowledge of the
behavior of ctenophores in the sea.
Acknowledgments
I thank James L. Olds for asking the question that initiated this investigation. I am grateful to Benjamin Moran and
Shannon Jones of the Bay Paul Center, MBL, for their
generous assistance in preparing column charts with error
bars; to David Mark Welch, Director of the Bay Paul
Center, for significant advice on statistics; to Susan Banks
of the Bell Center for her expert help in scanning figures;
and to Mark Terasaki for discussions and support. Signhild
Tamm spotted the ideal item to use for the statolith in the
model.
Literature Cited
Aronova, M. 1974. Electron microscopic observations on the aboral
organ of ctenophores. Z. Mikrosk. Anat. Forsch. 88: 401– 412.
Budelmann, B. U. 1988. Morphological diversity of equilibrium receptor systems in aquatic invertebrates. Pp. 757–782 in Sensory Biology of
Aquatic Animals, J. Atema, R. R. Fay, A. N. Popper, and W. N.
Tavolga, eds. Springer-Verlag, New York.
Chun, C. 1880. Die Ctenophoren des Golfes von Neapel und der angrenzenden Meeres-Abschnitte. Flora und Fauna des Golfes von Neapel, Vol. 1. Engelmann, Leipzig.
Crick, F. 1988. What Mad Pursuit. A Personal View of Scientific Discovery. Basic Books, New York. 182 Pp.
Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco
and the Panglossian paradigm: a critique of the adaptationist programme. Proc. R. Soc. Lond. B 205: 581–598.
Horridge, G. A. 1966. Pathways of co-ordination in ctenophores. Pp.
247–266 in The Cnidaria and Their Evolution, W. J. Rees, ed. Zool.
Soc. London 16. Academic Press, New York.
Horridge, G. A. 1971. Primitive examples of gravity receptors and their
evolution. Pp. 203–221 in Gravity and the Organism, S. A. Gordon and
M. J. Cohen, eds. University of Chicago Press, Chicago.
Horridge, G. A. 1974. Recent studies on the Ctenophora. Pp. 439 – 468
in Coelenterate Biology: Reviews and New Perspectives, L. Muscatine
and H. M. Lenhoff, eds. Academic Press, New York.
Krisch, B. 1973. Über das Apikalorgan (Statocyste) der Ctenophore
Pleurobrachia pileus. Z. Zellforsch. 142: 241–262.
Lowe, B. 1997. The role of calcium in deflection-induced excitation of
motile mechanoresponsive balancer cilia in the ctenophore statocyst. J.
Exp. Biol. 200: 1593–1606.
Moroney, M. J. 1956. Facts From Figures. Penguin Books, Baltimore,
MD.
Noda, N., and S. L. Tamm. 2014. Lithocytes are transported along the
ciliary surface to build the statolith of ctenophores. Curr. Biol. 24:
R951–R952.
Samassa, P. 1892. Zur Histologie der Ctenophoren. Arch. Mikrosk. Anat.
40: 157–243.
Schnitzler, C. E., K. Pang, M. L. Powers, A. M. Reitzel, J. F. Ryan, D.
Simmons, T. Tada, M. Park, J. Gupta, S. Y. Brooks et al. 2012.
Genomic organization, evolution, and expression of photoprotein and
opsin genes in Mnemiopsis leidyi: a new view of ctenophore photocytes. BMC Biol. 10: 107.
Tamm, S. L. 1980. Cilia and ctenophores. Oceanus 23: 50 –59.
Tamm, S. L. 1982. Ctenophora. Pp. 266 –358 in Electrical Conduction
and Behaviour in ‘Simple’ Invertebrates, G. A. B. Shelton, ed. Oxford
University Press, Oxford.
Tamm, S. L. 2012a. Regeneration of ciliary comb plates in the ctenophore Mnemiopsis leidyi. I. Morphology. J. Morphol. 273: 109 –120.
Tamm, S. L. 2012b. Patterns of comb row development in young and
adult stages of the ctenophores Mnemiopsis leidyi and Pleurobrachia
pileus. J. Morphol. 273: 1050 –1063.
Tamm, S. L. 2014a. Cilia and the life of ctenophores. Invertebr. Biol.
133: 1– 46.
Tamm, S. L. 2014b. Ctenophores and termites—systems for motility.
Pp. 147-171 in Cilia and Flagella: Ciliates and Flagellates, K. Hausmann and R. Radek, eds. Schweizerbart, Stuttgart, Germany.
Tamm, S. L. 2014c. Formation of the statolith in the ctenophore Mnemiopsis leidyi. Biol. Bull. 227: 7–18.
Tamm, S. L., and S. Tamm. 2002. Novel bridge of axon-like processes
of epithelial cells in the aboral sense organ of ctenophores. J. Morphol.
254: 99 –120.